Firm Organizational Environment and The Effect of Options ... · PDF...

Firm Organizational Environment and

The Effect of Options Based Compensation Incentives on

Accrual Earnings Management

Yacine Belghitar

1

Ephraim Clark2

Abstract

This paper explores the effect of firm organizational environment on

the relationship between the option based incentives of CEO compensation

and accrual earnings management. Results show that this relationship varies

according to the firm’s organizational environment with respect to growth

prospects, capital structure, and riskiness. Only delta and the change in delta

are significant for high growth and high leverage firms. For low growth and

low leverage firms only vega is significant. For low volatility firms delta,

change in delta and vega are all significant while none are significant for high

volatility firms. Our results have important implications for managers, policy

makers, and compensation boards.

Keywords: Earnings management, accruals, CEO compensation, utility

function, delta, change in delta, vega.

JEL Classification : F30; F32, G32, G33.

1 Cranfield School of Management, [email protected]

2 Middlesex University, [email protected]

mailto:[email protected]

mailto:[email protected]

Yacine Belghitar, Ephraim Clark – Firm Organizational Environment and

The Effect of Options Based Compensation Incentives on Accrual Earnings Management –

Frontiers in Finance and Economics – Vol 12 N°2, - 1-29

2

1 – Introduction

A growing literature examines the effect of equity based

compensation incentives, such as delta and vega,3on financial misreporting

via accrual earnings management. We extend the literature by showing that

the effect of these incentives on accrual earnings management is influenced in

significant ways by the firm’s organizational environment with respect to

growth opportunities, capital structure and risk profile.

Accrual earnings management is a well-documented fact of corporate

behaviour (e.g. Healey and Wahlen, 1999 and Cohen, Dey, and Lys, 2004). 4

While accrual earnings management can be a useful signalling tool to improve

information disclosure about the firm, to improve the ability of earnings to

reflect economic value, to dress up financial statements prior to public

securities offerings, to avoid violating lending contracts, to reduce regulatory

costs or increase regulatory benefits, it can also be a tool to increase corporate

managers’ compensation and job security at the expense of shareholders (see,

e.g. Bergstresser and Philippon, 2006).

Given the multiple potentially conflicting motives for accruals

management, the extant empirical literature is unsurprisingly mixed and

suggests that the nature of the relationship between option based incentives

(delta and vega) and accruals management depends crucially on how accruals

management affects the firm’s stock price and stock price volatility.5 For

example, Bergstrasser and Philippon (2006) and Cornett, Marcus and

Tehranian (2008) argue that accruals management increases the firm’s share

price and find a positive relation between CEO delta and accrual earnings

management while Jiang, Petroni and Wang (2010) find no relation for the

3Delta corresponds to the change in managerial wealth with respect to a change in the value of

the firm’s stock, while vega corresponds to the change in managerial wealth with respect to the

change in the volatility of the firm’s returns. 4 See Defond and Jiambalvo (1994), Rees et al. (1996), Teoh et al. (1998a), and Teoh et al.

(1998b) for the use of discretionary accruals in tests of earnings management and market

efficiency. 5The extant studies on the relationship between CEO option based incentives and financial

misreporting in general have also been inconclusive. For example, Burns and Kedia (2006) find

a positive relation between delta and restatements. Erickson, Hanlon, and Maydew (2006)find

no relation between delta and accounting fraud or misrepresentation while Feng, Ge, Luo and

Shevlin(2011) a positive relation for the CFO and no relation for the CEO. Armstrong,

Jagolinzer and Larcker (2010) find no evidence of a relation between delta and litigation,

restatements and accounting fraud or misrepresentation.




3

CEO and a positive relation for the CFO. In this context, the argument is that

a higher delta motivates a corporate manager to increase accruals management

in order to benefit from the increase in his equity portfolio as the share price

rises (the reward effect), but at the cost of increased exposure to the firm’s

total risk (the risk effect).6 Their results are joint evidence that over the

sample they study accruals management had the effect of increasing the share

price and that the reward effect was stronger than the risk effect.

Chava and Purnanandam (2010) assume that accruals management

increases the stock price and decreases stock price volatility. They find a

positive relation for delta and a negative relation for vega. The positive

relation for delta is evidence that over the sample they study accruals

management had the effect of increasing the share price and that the reward

effect was stronger than the risk effect. The negative relationship between

vega and accruals management is joint evidence that accruals management

reduced share price volatility and that corporate managers reduce accrual

management in order to avoid the loss to their equity portfolios that would be

caused by a reduction in the share price volatility.

Armstrong et al. (2013) argue that accruals management increases

both the share price and share price volatility. They find a positive relation for

delta and a positive relation for vega. The positive relation for delta is

evidence that over the sample they study accruals management had the effect

of increasing the share price and that the reward effect was stronger than the

risk effect.7 The negative relationship between vega and accruals management

is joint evidence that accruals management increased share price volatility

over the sample they study and that corporate managers increase accrual

management in order to benefit from the increase in their equity portfolios due

to the increased volatility.

There is, however, no definitive theoretical or empirical evidence that

accruals management systematically increases the firm’s share price. For

example, Sloan (1996) and Kothari et al. (2011) find that accruals

management actually reduces the firm’s share price. 8

In this situation, the

6 Unlike a well diversified outside shareholder, managers typically hold disproportionately

higher fraction of their wealth in the firm. In addition, managers’ human capital is closely tied

with the firm’s performance (see Fama, 1980; Stulz, 1984; Smith and Stulz, 1985). Given these

considerations, managers with higher delta are likely to prefer low risk financial policies to

minimize the firm’s total risk. 7 When vega and delta appear together in the testing, this effect is subsumed by vega. 8There is substantial evidence in the literature that earnings management can reduce firm

performance. For example, Huanga, et al. (2009) show that earnings management is inversely




4

relationship between accruals management and is reversed. A higher delta

will discourage accrual management because the value of a manager’s equity

portfolio decreases as the stock price decreases. On the other hand, it will

encourage accrual management because delta amplifies the effect of equity

risk on the total riskiness of a manager’s equity portfolio.

Ultimately, the relationship between accruals management and option

based compensation incentives is an empirical question that depends crucially

on the effect that managing the accruals has on the share price. The effect on

firm share price could vary from one sample to another depending on the

prevailing economic and financial environment. It could also vary from one

firm to another within the same sample depending on the firm’s

organizational environment.9 For example, the growth opportunity set is a key

element in the organizational environment. Firms with greater growth

opportunities are subject to greater levels of informational asymmetry and

agency costs (Smith and Watts, 1992 and Belghitar and Clark, 2011).

Accruals management for high growth firms could be used as a signalling tool

to overcome the informational asymmetry. However, firms with higher

growth opportunities are also subject to higher bankruptcy costs (Harris and

Raviv, 1990; Shleifer and Vishny, 1992). Shin and Stulz (2000) argue that

since firm risk and growth opportunities may be positively linked,

undiversified managers in high growth firms may have greater incentive to be

risk averse than those in low growth firms. Thus, high growth and low growth

firms differ in the motivation for accruals management and how it affects firm

share priceas well as the degree of managerial risk aversion.

Another key element in the firm’s organizational environment that

can affect the effectiveness of managerial compensation incentives is the

firm’s capital structure. Under the bondholder wealth expropriation

hypothesis developed by Black and Scholes (1973), unanticipated increases in

the riskiness of the firm’s assets transfer wealth from bondholders to

shareholders. Based on this, and other things being equal, shareholders have

an incentive to increase the riskiness of the firm’s assets and bondholders

have an incentive to reduce it.Within certain boundaries (see, for example,

Jensen and Meckling, 1976) shareholders also have an incentive to increase

leverage, which also increases the riskiness of equity (as opposed to the

related to firm value. In a similar vein, Aboody et al. (2005) and Francis et al.(2005) show that

accrual earning management leads to higher costs of capital. DeFond and Park (1997) show that

firms use accrual management to understate the volatility of earnings. 9Hutchinson and Gul (2004) have shown that this is the case with other corporate governance

mechanisms.




5

riskiness of the firm’s total assets). To protect themselves against these

sources of “wealth expropriation” bondholders often impose covenants that

restrict corporate borrowing and risk-taking. Thus, different levels of leverage

create different risk dynamics for managerial compensation incentives. For

example, because of restrictive covenants at higher levels of

leverage,vegamight be a less effective incentive than delta to affect accruals

management.

Finally, the level of firm volatility could also affect the impact of

managerial compensation incentives on earnings management. The argument

here is that the incentive to increase risk can hold only if the level of risk

aversion is compatible with risk higher than the ongoing level of risk. If it is

higher, the tendency will be to reduce volatility. Thus, if firm volatility is high

enough, it is conceivablethat no incentive scheme would lower the level of

risk aversion enough to make the manager increase the volatility. Conversely,

if firm volatility is low enough,option based incentives would have more

scope for increasing volatility.

Another innovation of this paper is that we build our research design

around the concepts of convexity, magnification and translation10

in Ross

(2004) operationalized by Belghitar and Clark (2014) with respect to the

Black/Scholes option pricer. Belghitar and Clark (2014) show that the

relationship between option based compensation incentives and risk aversion

depends on whether the managerial utility function shows decreasing,

increasing or constant absolute risk aversion (DARA, IARA, CARA

respectively).11

More specifically, they identify two parameters, delta and the

change in delta, that appear directly in the Arrow-Pratt measure of CEO risk

aversion.12

The change in delta corresponds to the “risk effect” in Armstrong

et al. (2013). It increases CEO risk aversion through its effect on convexity

and magnification. The effect of delta itself corresponds to the “reward effect”

in Armstrong et al. (2013) and depends on whether the CEO has DARA,

IARA or CARA. Belghitar and Clark (2014) also show that although vega

does not appear directly in the measure of risk aversion, it could impact risk

10The convexity effect is how the individual option based incentives affect managerial risk

aversion at a given level of wealth. The magnification effect reflects how different levels of

exposure to the firm’s share price affect managerial risk aversion. The translation effect is how

different levels of wealth affect managerial risk aversion.

11 See Appendix for details. 12Abdel-Khalik (2007) establish an empirical link between CEO risk aversion and earnings

management.




6

aversion through its effect on CEO wealth. As with delta, the effect of vega

depends on whether the CEO has DARA, IARA or CARA.

In the main contribution to the literature, we show that although the

relationship between the option based incentives and accrual earnings

management is strong and significant, it varies according to the firm’s growth

prospects, its capital structure and overall firm riskiness. Only delta and the

change in delta are significant determinants of abnormal accruals in high

growth firms, while only vega is significant in low growth firms. The

relationship between firm based compensation and abnormal accruals

disappears in high leverage firms where the CEO/shareholder agency conflict

is neutralized by the shareholder/bondholder agency conflict. It also

disappears in high risk firms where the agency conflict between CEO and

shareholders is not apparent. Our results provide an explanation for some of

the apparent contradictions in the outstanding literature and have important

implications for corporate compensation policies and accounting regulations.

The rest of this study is structured as follows. Section 2 provides a

description of the sample utilized in the current study and develops the links

between delta, the change in delta and vega and accrual earnings management.

Section 3 presents the empirical results. Section 4 concludes.

2 – Data and methodology

The initial sample consists of all UK firms that are listed on the 350

FTSE as of January 2000. Data on managerial compensation sensitivities and

data regarding the number of directors and the number of independent non-

executive directors on the board are obtained from BoardEx. Company

balance sheet information used for estimating abnormal accruals and other

control variables used for the analysis are collected from Thomson Reuters

Worldscope. After excluding financial firms and firms without the requisite

information on managerial compensation sensitivities, the final sample

consists of an unbalanced sample of 270 firms for the period 2000 through

2004. 13

13

We settled on this time period because of data availability and because it was

homogenous in so far as it was pre-crisis period and relatively free of potential

distortions due to major tax reforms. The year 2000 is the first year that Boardex

reports estimations of CEO vega and delta. It is also the first year after the end of the

effects of UK tax reform of July 1997 on equity prices as reported by Bell and




7

2.1 - Measurement of Earnings Management

In this study we use abnormal accruals as our proxy for accrual

earnings management. To this end, we follow closely Dechow et al. (1995),

Bergstresser and Philippon (2006) and Caramanis and Lennox (2008) to

define total accruals as:

Where is total accruals for firm i at time t; ∆ is the first

difference operator for a one year change; is change in current assets

for firm i at time t; is the change in current liabilities of firm i at time

t; is the change in cash and short-term investments for firm i at time t;

is the change in short-term debt and current portion of long-term

debt for firm i at time t; is the depreciation and amortisation expense

for firm i at time t; and is the lagged assets of firm i at time t-1.

To obtain abnormal accruals, we use the performance matched model

advocated by Kothari et al. (2005) and estimate equation (2.2) cross-

sectionally for each year using all firm-year available observations in the

same industry code:14

Jenkinson (2002). Boardex modified its database for the post-2004 period due to the passage of FAS 123R on December 12, 2004. Up to 2005 Boardex reports compensation data

using the old format (pre-FAS 123R). For fiscal years 2005 and later, Boardex reports

compensation using the new format (post-FAS 123R). In the post-FAS 123R period, firms

calculate and expense equity-based compensation at fair value using their own valuation

models. Thus, for the post-2004 data, Boardex does not calculate the Black-Scholes value of

current year stock option grants, nor do they provide estimates of CEO vega and delta. Instead,

Boardex reports the firm‘s own calculated fair values of equity-based compensation, which is

not comparable across firms within the same year if firms are using different valuation

methods. Additionally, for the same firm, CEO vega and deltaare not comparable pre- and post-

FAS 123R.

14As a robustness check and comparison with other studies, we also estimate equation 2.2

without the constant term. To justify the constant, Kothari et al. (2005) p. 173 writes: “While

prior research typically does not include a constant in the above model, we include a constant in

the estimation for several reasons. First, it provides an additional control for heteroskedasticity

not alleviated by using assets as the deflator. Second, it mitigates problems stemming from an

omitted size (scale) variable.Finally, we find that discretionary accrual measures based on

models without a constant term are less symmetric, making power of the test comparisons less




8

where is total accrual for firm i at time t; ∆ is the first difference

operator for a one year change; is the slope intercept; is the i-th slope

coefficient for i {1,2,3}; is the lagged assets of firm i at time t-1;

is the change in revenue for firm i at time t, scaled by the lagged value

of assets; is the plant, property and equipment for firm i at time t scaled

by the lagged value of assets; is return on assets for firm i at time t; is the stochastic error term. Subsequent to the estimation of equation 2.2, the

absolute value of the regressions residual term is utilised as a proxy for the

magnitude and propensity for abnormal accruals and by extension accrual

earnings management.

Since accruals management depends crucially on how accruals

management affects the firm’s stock price and stock price volatility, in results

not reported here but available on request we look at the relationship between

lagged abnormal accruals and market returns to equity as well as the volatility

of market returns to equity. The correlation coefficient for both is negative

and significant. OLS regressions of market returns to equity and the volatility

of market returns to equity on lagged abnormal returns with and without

control variables confirm the significant negative relationship. This situates

our overall sample in the zone where the overall effect of accruals on stock

market returns and return volatility is negative.

2.2 - Managerial Compensation Incentives

Following Belghitar and Clark (2014), we use as CEO option based

risk incentives the delta, the change in delta and the vega of CEO firm based

wealth.15

Firm based wealth includes the value of all stock ownership,

clear-cut. Thus, model estimations including a constant term allow us to better address the

power of the test issues that are central to our analysis”.

15The delta of a share is equal to 1 and its vega is equal to 0. The delta and vegaof total firm

based wealth is the weighted average of the deltas and vegas of the individual shareholdings

and options: i

iiz and i

ii vegazvega where i refers to the individual




9

unexpired stock options and long term incentive plans (LTIPs) accumulated

and held by the CEO to date. CEO delta (CEO_Delta) measures the pound

gain in the CEO’s firm based wealth following a stock price increase of 1%.

The change in CEO delta (CEO_DeltaChg) is captured by computing the first

difference of delta [CEO_Delta time T - CEO_Delta time T-1]. CEO vega is

defined as the pound gain in the CEO’s firm based wealth as stock return

volatility increases by 1%.16

Delta acts on risk aversion through the translation effect derived in

Ross (2004). Its effect depends on whether the manager’s utility function has

decreasing, increasing or constant absolute risk aversion.17

The effect on risk

aversion is negative with DARA.18

It is positive with IARA19

and it is equal to

zero with CARA.20

This implies that if the CEO is using accrual earnings

management to aggressively manipulate earnings for his own personal benefit

at the expense of shareholders, the effect of delta on abnormal accruals will

depend on the CEO utility function. Thus, we have no prior on the sign of

delta.

Changes in delta affect risk aversion through their effect on convexity

and magnification. An increase in delta will reduce the effects of convexity

shareholdings and options and iz is the proportion of asset i in total firm based wealth. For

example, consider a manager with 50% of his firm based wealth in shares and 50% in an option

with a delta of 0.5. The delta of his portfolio will be equal to 75.05.05.015.0 .

. 16All compensation sensitivities are scaled by thousands of pounds similar to previous studies

(Guay, 1999; Coles, et al, 2006). 17Ross calls this moving the evaluation to a different part of the domain of the original utility

function where the utility function can have greater or lesser risk aversion.

18 Let A represent absolute risk aversion, w represent wealth and u the CEO utility function,

where primes signify first and second derivatives:

DARA implies 0)(

)( -

wu

wu

dw

dA

19IARA implies 0)(

)( -

wu

wu

dw

dA

20CARA implies 0)(

)( -

wu

wu

dw

dA




10

and magnification and increase managerial risk aversion. This implies that if

the CEO is using accrual management to manipulate reported earnings for his

own personal benefit at the expense of shareholders, an increase in CEO delta

will increase abnormal accruals. Thus, we expect a positive relationship

between change in delta and abnormal accruals.

Vega does not appear directly as a determinant of CEO risk aversion

in Ross (2004). It could, however, affect risk aversion indirectly through the

effect of volatility on CEO wealth. It is well known in the option pricing

literature that the first derivative of the option price with respect to the

volatility of the firm’s returns/earnings to equity is positive. An increase in

volatility increases the CEOs firm based wealth.21

The higher the vega, the

larger is the increase in wealth. An increase in wealth affects risk aversion

depending on whether the manager has decreasing, increasing or constant

absolute risk aversion. As with delta, the effect is negative with DARA,

positive with IARA and equal to zero with CARA. Thus, higher levels of vega

should increase the incentive of CEOs with DARA to reduce abnormal

accruals. Higher levels of vega should increase the incentive of CEOs with

IARA to increase abnormal accruals. Higher levels of vega should have no

effect on abnormal accruals for CEOs with CARA.Sincevega and delta affect

risk aversion in a similar manner, they should both have the same sign with

respect to abnormal accruals.

2.3 - Control Variables

The control variables employed in this study are all consistent with

prior studies on accrual earnings management. Following Klein

(2002)andPeasnell et al. (2005), we control for board size and the percentage

of independent non-executive directors on the board where the natural

logarithm of total directors on the board, and the number of independent non-

executives to total directors on the board are used as respective proxies. We

include CEO cash compensation, measured as the sum of all cash based

compensation received by the CEO during the year. We use the audit fees as a

proxy for audit effort.Wealsocontrol for the following firm characteristics:

21Other things being equal, the vega of a share is equal to zero. Like delta, the vega of

managerial total wealth is a weighted average of the vegas of the individual shareholdings and

options: i

ii vegazvega .




11

leverage and firm size. Leverage is defined as the ratio of long-term debt to

total assets. The natural logarithm of total sales is utilized as a proxy for firm

size. Finally, the ratio of research and development plus selling, general and

administrative expenses to total assets is used for firm uniqueness similar to

Titman and Wessells (1988). Descriptive statistics for the main variables are

presented in Table 1.

Table 1: Descriptive Statistics

Variable Mean Median SD

AB_ACC 0.164 0.059 0.337

AB_ACC2 0.146 0.062 0.303

CEO_Delta 82.042 22.000 198.192

CEO_Vega 17.171 1.253 99.674

CEO_DeltaChg -3.512 3.000 182.362

BSIZE 2.085 2.079 0.304

CEO_Cash 516.390 395.6 444.014

NEXEC 0.436 0.429 0.158

AUDIT 6.378 6.217 1.455

R&D 0.283 0.206 0.298

LEV 0.165 0.130 0.164

RISK 0.380 0.321 0.222

SIZE 12.696 12.922 2.155

AB_ACC is the absolute value of residuals obtained from equation 2.3.

AB_ACC2 is the absolute value of residuals obtained from equation 2.2

excluding the constant term. CEO_Delta is the pound change in CEO wealth

for a 1% change in stock price. CEO_Vega is the pound change in the CEO’s

wealth for a 0.01 change in stock return standard deviation. CEO_DeltaChg

is the first change in CEO_Delta. CEO_Cash is measured as the sum of all

cash based compensation received by the CEO during the year in

thousands.Bsize is the natural logarithm of total directors on the board.

NEXEC is the ratio of total independent non-executive directors on the board

to total directors on the board. AUDIT is the natural logarithm of audit fees.

R&D is the ratio of research and development plus selling, administrative and

general expenses to total assets. LEV is the ratio of long-term debt to total

assets. RISK is the annualised standard deviation of year t daily stock return.

SIZE is the natural logarithm of total sales




12

3 – Empirical results

To examine the impact of CEO compensation incentives on earnings

management we employ pooled, multivariate regression analysis controlling

for both industry-effects and time-effects. One salient issue is the potential

endogeneitybetween earnings management and its hypothesized determinants.

The independent variables are lagged by one year in order to reduce the

potential endogeneity bias. In regression 1, the dependent variable, absolute

abnormal accruals (noted AB_ACC), is estimated using equation (2.1), while

in regression 2the dependent variable (noted AB_ACC2) is estimated using

equation (2.2) without a constant. The regression results are reported in Table 2.

Table 2: Determinants of Earnings management (1) (2)

CEO_Delta -0.00020** -0.00022**

(-2.82) (-2.98)

CEO_Vega -0.00008** -0.00009**

(-2.63) (-2.85)

CEO_DeltaChg 0.00009* 0.00010*

(1.84) (1.81)

CEO_Cash -0.18955** -0.20977*

(-1.97) (-1.87)

BSIZE 0.07802 0.06814

(0.71) (0.57)

NEXEC -0.11860 -0.15238

(-0.95) (-1.07)

AUDIT -0.00912 0.01183

(-0.62) (0.73)

RD -0.04761 -0.06113

(-0.92) (-1.12)

LEV 0.30741** 0.42498**

(2.43) (2.88)

RISK -0.07522 -0.12833

(-0.66) (-1.02)

SIZE -0.00155 -0.00798

(-0.22) (-1.01)

Constant 0.57038** 0.62635**

(4.13) (4.02)

Ind. dummies Yes Yes

Year dummies Yes Yes

F 3.99** 3.72**

R2 3.81% 4.88%




13

The dependent variables are AB_ACC and AB_ACC2 in models (1) and (2) are

measured as the absolute value of residuals obtained from equation 2.2 with and

without the constant term respectively. CEO_Delta is the pound change in CEO

wealth for a 1% change in stock price. CEO_Vega is the pound change in the

CEO’s wealth for a 0.01 change in stock return standard deviation. CEO_DeltaChg

is the first change in CEO_Delta.CEO_Cash is measured as the natural logarithm of

the sum of all cash based compensation received by the CEO during the year.BSIZE

is the natural logarithm of total directors on the board. NEXEC is the ratio of total

independent non-executive directors on the board to total directors on the board.

AUDIT is the natural logarithm of audit fees. R&D is the ratio of research and

development plus selling, administrative and general expenses to total assets. LEV is

the ratio of long-term debt to total assets. RISK is the annualised standard deviation

of year t daily stock return. SIZE is the natural logarithm of total sales. The

coefficients’ standard errors are adjusted for the effects of non-independence by

clustering on each firm (Petersen 2009). The number of observations in the

regression is 698. t statistics in parentheses. ** and * indicate statistical significance

at the 5% and 10% level respectively

Our results provide evidence that managerial compensation incentives

play an important role in influencing earnings management decisions. All

three incentive measures, CEO delta, the change in CEO delta

(CEO_DeltaChg) and CEO vega, are significant and have the signs suggested

in section 2. CEO delta and vega have the same negative sign, which is

evidence that the CEOs in our sample have DARA.22

The change in CEO

delta is positive, meaning that larger changes in delta increase earnings

management. CEO Cash is negative and significant. Overall our findings

suggest that managerial compensation incentives exert significant influence

on the decision to employ accruals to manage earnings.

With respect to firm characteristics, leverage is positive and

significant in both regressions. This is consistent with the findings of Klien

(2002),Bergstresser and Philippon (2006), and Caramanis and Lennox (2008).

None of the other firm specific variables are significant at any conventional

level. The fact that none of the governance variables, such as board size,

independent board members and auditing cost is significant is evidence of the

wide latitude that CEOs have in the decision to allocate and reprise certain

types of accruals.

22This substantiates Abdel-Khalik’s (2007) assumption of DARA.




14

3.1 - The Growth Opportunity Set, Earnings Management and

Managerial Compensation Sensitivities

As argued in the introduction, the effectiveness of managerial

compensation incentives may depend on the organisational environment in

which a firm operates as is the case with other corporate governance

mechanisms (Hutchinson and Gul, 2004). The growth opportunity set is a key

element in the organizational environment. It is generally accepted that firms

with greater growth opportunities are subject to greater levels of informational

asymmetry and agency costs (Smith and Watts, 1992; Belghitar, Clark and

Kassimatis, 2011). Firms with higher growth opportunities are also subject to

higher bankruptcy costs (Harris and Raviv, 1990; Shleifer and Vishny, 1992).

Shin and Stulz (2000) argue that since firm risk and growth opportunities may

be positively linked,undiversified managers in high growth firms may have

greater incentive to be risk averse than those in low growth firms. Therefore,

we examine the impact of managerial compensation incentives in influencing

earnings management in firms with large and small growth opportunity sets.

To investigate the impact of managerial compensation sensitivity in

influencing earnings management in different growth environments we utilise

the ratio of market value of equity to book value of equityas a proxy for

growth opportunities.23

We split our sample into three categories. Firms with

growth opportunity less than the 40th percentile are classified as small growth

opportunity firms and firms with growth opportunity greater than the 60thare

categorised as large growth firms. The results are presented in Table 3.

Models (3) and (4) refer to regressions for high and low growth firms where

the dependent variable is AB_ACC. Models (5) and (6) refer to high growth

and low growth regressions where the dependent variable is AB_ACC2.

In both regressions for high growth firms (models (3) and (5)) both

CEO_Delta and CEO_DeltaChgare significant and have the predicted signs.

Delta is negative, which is evidence for DARA, and CEO_DeltaChgispositive,

which is evidence that the sensitivity to the convexity and magnification

effects decreases as delta increases. Vega, however, is not significantwiththe

implication that managers in high growth firms are wary of firm risk, even

when it can increase their wealth. This is evidence for the Shin and Stulz

(2000) argument that undiversified managers in high growth firms may have

greater incentive to be risk averse than those in low growth firms.

23

As a robustness test we also use Tobin’s Q to proxy for growth opportunities and get weaker

but qualitatively similar results.




15

In both regressions for low growth firms (models (4) and (6)) vega is

negative and significant, which is evidence of DARA formanagers in low

growth firms.Delta is also negative but not significant. Remember that delta

reflects the sensitivity of managerial wealth to changes in the firm share price.

Low growth firms have low prospects for gains in the share price. Thus, the

strong effect of vega and the insignificant effect of delta might reflect the fact

that CEOs of low growth firms see wealth enhancement through increases in

volatility as more likely than through increases in the firm’s share price. The

fact that CEO_DeltaChg is not significant reinforces this argument. The

convexity and magnification effects that depend on the value of the firm share

price may not matter much if the share price is not likely to change much.

Table 3: Earnings management and investment opportunity growth High

growth

Low

growth

T-diff

High

growth

Low

growth

T-diff

(3) (4) (5) (6)

CEO_Delta -0.00022** -0.00025 2.14** -0.00038** -0.00024 2.02**

(-2.92) (-1.39) (-3.71) (-1.29)

CEO_Vega 0.00053 -0.00006** 2.05** 0.00072 -0.00007* 1.92*

(1.13) (-2.03) (1.18) (-1.90)

CEO_DeltaChg 0.00011* 0.00017 1.85* 0.00020** 0.00016 2.19**

(1.93) (1.09) (3.17) (1.03)

CEO_Cash -0.02455 0.04925 1.55 -0.05657 0.01606 1.43

(-0.67) (0.62) (-1.21) (0.20)

BSIZE 0.11244 -0.38055 0.46068 -0.18944

(0.34) (-0.62) (1.18) (-0.29)

NEXEC -0.18842 0.00183 -0.19318 -0.07687

(-1.02) (0.01) (-0.86) (-0.29)

AUDIT -0.02304 -0.02878 -0.01885 0.00876

(-1.16) (-0.94) (-0.76) (0.25)

RD -0.02863 -0.21183 -0.06294 -0.17918

(-0.63) (-1.40) (-1.21) (-1.20)

LEV 0.28301 0.33453* 0.07448 0.59111**

(1.54) (1.94) (0.36) (2.79)

RISK -0.12398 0.11451 -0.13821* 0.03713

(-1.59) (0.52) (-1.74) (0.15)

SIZE 0.00848 -0.02298 0.01546 -0.04738

(1.07) (-0.83) (1.63) (-1.52)

Constant 0.30719 1.02225 -0.17633 0.99979

(0.77) (1.35) (-0.41) (1.24)

Ind. dummies Yes Yes Yes Yes

Year dummies Yes Yes Yes Yes

F-test 1.93** 2.1** 2.13** 1.88**

R2 6.27% 5.47% 6.21% 8.33%

The dependent variables in models (3) and (4) are AB_ACC and AB_ACC2 in models (5) and




16

(6) are measured as the absolute value of residuals obtained from equation 2.2 with and without

the constant term respectively. CEO_Delta is the pound change in CEO wealth for a 1% change

in stock price. CEO_Vega is the pound change in the CEO’s wealth for a 0.01 change in stock

return standard deviation. CEO_DeltaChg is the first change in CEO_Delta. CEO_Cash is

measured as the natural logarithm of the sum of all cash based compensation received by the

CEO during the year. BSIZE is the natural logarithm of total directors on the board. NEXEC is

the ratio of total independent non-executive directors on the board to total directors on the

board. AUDIT is the natural logarithm of audit fees. R&D is the ratio of research and

development plus selling, administrative and general expenses to total assets. LEV is the ratio

of long-term debt to total assets. RISK is the annualised standard deviation of year t daily stock

return. SIZE is the natural logarithm of total sales. The coefficients’ standard errors are

adjusted for the effects of non-independence by clustering on each firm (Petersen 2009). T-Diff

is the t-test of difference between regression coefficients. The number of observations in the

regressions for testing the difference in the coefficients is 698. t statistics in parentheses. ** and

* indicate statistical significance at the 5% and 10% level respective.

3.2 - Leverage, Earnings Management and Managerial Compensation

Sensitivities

Another key element in the firm’s organizational environment that

lends itself to assessing the effectiveness of managerial compensation

incentives is the firm’s capital structure. As argued in the introduction,

covenants associated with the bondholder wealth expropriation hypothesis can

restrict corporate borrowing and risk-taking. Thus, different levels of leverage

create different risk dynamics for managerial compensation incentives.

Therefore, we examine the impact of managerial compensation incentives in

influencing earnings management in firms with large and small leverage ratios.

To investigate the impact of managerial compensation incentives in

influencing earnings management at different leverage ratios, we proceed as

above and split our sample into 3 categories. Firms with leverage ratios less

than the 40th percentile are classified as low leverage firms and firms with

leverage ratios greater than the 60th are categorized as high leverage firms.

The results of the tests using the same methodology as above are presented in

Table 4. Models (7) and (8) refer to regressions for high and low leverage

firms where the dependent variable is AB_ACC. Models (9) and (10) refer to

high and low leverage firms where the dependent variable is AB_ACC2.

The results are similar to those in table 3. In both regressions for high

leverage firms (models (7) and (9)) CEO_Delta and CEO_DeltaChgare

significant and have the predicted signs. Delta is negative, which is evidence

for DARA, and CEO_DeltaChg is positive, which is evidence of sensitivity to

the convexity and magnification effects. Vega, however, is not significant. As




17

in the case of managers in high growthfirms, this implies that managers in

high leverage firmswhere volatility is already high due to the leverageare

wary of more firm risk, even when it can increase their wealth. This is

evidence that undiversified managers in high leverage firms may have greater

incentive to be risk averse than those in low leverage firms.

In both regressions for low leverage firms (models (8) and (10)) vega

is negative and significant, which is evidence of DARA for managers in low

leverage firms. Delta is also negative but not significant. Since, ceteris

paribus, lower leverage implies more latitude with respect to bondholder

restrictions and lower volatility of market returns, the strong effect of vega

and the insignificant effect of delta might reflect the fact that there is more

scope for CEO wealth creation from increased return volatility than from

increased share price. The fact that CEO_DeltaChg is not significant is

evidence that in a low leverage/low volatility environment, increases in

exposure to the firm’s share price are not a major determinant of CEO risk

aversion and earnings management.

Table 4: Earnings management and leverage High

leverage

Low

leverage

T-Diff

High

leverage

Low

leverage

T-Diff

(7) (8) (9) (10)

CEO_Delta -0.00032** -0.00008 1.98* -0.0003** -0.00009 2.15**

(-3.4) (-0.68) (-2.1) (-0.90)

CEO_Vega -0.00003 -0.00012** 1.15 -0.00003 -0.00016** 1.79*

(-0.60) (-2.60) (-0.49) (-3.25)

CEO_DeltaChg 0.00012* 0.00001 2.27** 0.00009 0.00004 2.32**

(1.80) (0.08) (2.58)** (0.43)

CEO_Cash 0.01150 0.00344 1.21 0.01529 0.01200 1.03

(0.12) (0.09) (0.16) (0.28)

BSIZE -0.28310 -0.04918 -0.44561 -0.17193

(-0.35) (-0.14) (-0.50) (-0.44)

NEXEC -0.29376 0.07940 -0.26249 0.00569

(-0.96) (0.61) (-0.79) (0.04)

AUDIT 0.01912 -0.04391** 0.01971 -0.01807

(0.68) (-2.09) (0.63) (-0.78)

RD -0.10201 -0.00265 -0.19815 0.01427

(-0.47) (-0.06) (-0.88) (0.35)

LEV 0.00838 0.15695 0.10857 0.43740

(0.04) (0.50) (0.40) (1.26)

RISK 0.01410 -0.12073* -0.05079 -0.13558*

(0.05) (-1.72) (-0.15) (-1.83)

SIZE -0.04270 0.01488** -0.03657 0.00908

(-1.38) (2.15) (-1.12) (1.31)

Constant 1.38483 0.32488 1.65086 0.43568




18

(1.29) (0.72) (1.41) (0.87)



F-test 1.77* 2.81** 1.94** 2.75**

R2 5.40% 5.96 5.48% 5.82%















regressions for testing the difference in the coefficients is 698. t statistics in parentheses ** and

* indicate statistical significance at the 5% and 10% level respectively.

3.3 - Risk, Earnings Management and Managerial Compensation

Sensitivities

As a final test to our empirical analysis, we examine the impact of

managerial compensation incentives in influencing earnings management at

different levels of firm risk, measured as the annualized standard deviation of

the daily stock returns. The argument here is that if firm volatility is high

enough, it is conceivable that no incentive scheme would lower the level of

risk aversion enough to make the manager increase the volatility. Conversely,

if firm volatility is low enough, option based incentives would have more

scope for increasing volatility.

As in the previous two sub-sections, we split our sample into 3

categories. Firms with risk levels less than the 40th percentile are classified as

low risk firms and firms with risk levels greater than the 60th are categorized

as high Risk firms. The results of the tests are presented in Table 5. Models

(11) and (12) refer to regressions for high and low risk firms where the

dependent variable is AB_ACC. Models (13) and (14) refer to high and low

risk firms where the dependent variable is AB_ACC2.

Interestingly, option based compensation incentives have no

significant effect on high risk firms. They affect only low risk firms (models




19

(12) and (14))wherethe change in CEO delta is positive and significant,

andCEO delta and CEO vega are both negative and significant. This suggests

a very limited scope for incentivizing CEO behavior with option based

compensation packages.

Table 5 Earnings management and firm risk High

risk

Low risk High risk Low risk

(11) (12) T-Diff (13) (14) T-Diff

CEO_Delta 0.00003 -0.00038** 2.02** -0.00014 -0.00044** 1.77*

(0.13) (-2.88) (-0.71) (-3.05)

CEO_Vega -0.00001 -0.00012** 1.16 -0.00002 -0.00012* 1.22

(-0.26) (-2.10) (-0.76) (-1.88)

CEO_DeltaChg -0.00005 0.00013** 1.67* 0.00003 0.00015** 1.86**

(-0.36) (2.41) (0.23) (2.17)

CEO_Cash -0.01386 0.00266 1.12 -0.04273** 0.02847 0.92

(-0.77) (0.03) (-2.00) (0.32)

BSIZE 0.02079 -0.08462 0.24884 -0.28825

(0.12) (-0.13) (1.29) (-0.41)

NEXEC -0.15113 -0.11058 -0.34335** -0.12448

(-1.10) (-0.47) (-2.05) (-0.48)

AUDIT -0.00702 -0.00824 0.00207 0.03125

(-0.38) (-0.31) (0.12) (1.05)

RD 0.03071 0.01976 -0.00459 0.06277

(0.55) (0.16) (-0.09) (0.48)

LEV 0.22469 0.37688* 0.22345 0.53262**

(1.20) (1.72) (1.08) (2.17)

RISK 0.21616 -0.22071 0.21903 -0.46273

(1.41) (-0.68) (1.30) (-1.29)

SIZE 0.00524 -0.01035 0.00324 -0.02435

(0.73) (-0.62) (0.54) (-1.34)

Constant 0.05300 0.67462 -0.10042 0.89018

(0.24) (0.89) (-0.43) (1.08)



F 1.79** 3.05** 1.65* 3.70**

R2 6.98% 3.93% 8.40% 4.97%








20











regressions for testing the difference in the coefficients is 698. t statistics in parentheses. ** and

* indicate statistical significance at the 5% and 10% level respectively.

4 – Summary and conclusion

Our findings suggest that option based compensation incentives do

have an effect on CEO risk aversion and accrual earnings management, but

that the effect varies with the organizational environment of the firm. In the

overall sample,delta, vega and the change in deltaare significant determinants

of earnings management. Higher levels of delta and vegareduceearnings

management, which is evidence that CEOs have decreasing absolute risk

aversion, and higher levels of the change in delta increase it.

The picture is more nuanced when different organizational

environments are considered. For high growth firms,CEO_Delta and

CEO_DeltaChgare significant and have the predicted signs. Vega, however, is

not significant. For low growth firms,vega is significant but CEO_Delta and

CEO_DeltaChgare not. Similarly, for high leverage firms, CEO_Delta and

CEO_DeltaChgare significant and have the predicted signs, whilevega is not

significant. For low leverage firms, vega is significant but CEO_Delta and

CEO_DeltaChg are not. Thus, it looks like the CEOs are reacting to the

option based incentives with respect to the opportunities and constraints

specific to each environment. It is particularly interesting that vega is only

significant in the lower risk environments associated with low growth/low

leverage where delta and the change in delta are not significant. This could be

a problem because contrary to delta and the change in delta, which are related

to the firm’s share price, vega is related only to risk. Increased risk for its own

sake is not necessarily in the shareholders interest unless it is associated with

higher NPV projects that add value to the firm.

When we look at the effect of option based incentives in an

environment of firm based risk, option based compensation incentives have




21

no significant effect on high risk firms. They affect only low risk firms

(models (12) and (14)) where the change in CEO delta is positive and

significant CEO delta and CEO vega are both negative and significant. This

suggests a very limited scope for high risk firms to incentivize CEO behaviour

with option based compensation packages.

The implication for compensation boards is that structuring option

based compensation packages is much more complicated than previous

literature suggests.24

A successful structure that reduces accrual management

depends on the firm’s organizational environment as well as the CEO’s

attitude towards risk.For high growth and high leverage firms, the change in

delta increases abnormal accruals and suggests that the option should be

priced far in the money or far out of the money where the changes in delta are

smallest. Far in the money options will be more effective in reducing

abnormal accruals for CEOs with DARA (higher delta reduces abnormal

accruals for DARA utility functions), while far out of the money options will

be more effective for those with IARA (higher delta increases abnormal

accruals for IARA utility functions). Since vega is significant but neither delta

nor the change in delta are significant for low growth and low leverage firms,

the option can be structured to make vega as large as possible as far out of the

money as possible in order to get the maximum vega effect at the lowest

option cost. For high risk firms, option incentives do not seem to be an

effective tool for reducing accrual management. On the other hand, for low

risk firms all three incentives are significant. Thus, we want the option to be

far in or out of the money in order to have a low change in delta. The structure

of delta and vega depends on whether the CEO has DARA, IARA or CARA.

Appendix

Following Ross (2004), each manager has a utility function )(wu

satisfying the following conditions:

wwuwu ,0)( ,0)( (A1)

where primes denote first and second derivatives with respect to wealth, noted

as w. The degree of absolute risk aversion (Pratt, 1964) is defined as

)(

)(

wu

wuA

(A2)

24 The structure can be modified by varying the strike price and time to maturity.




22

where A measures how much the manager dislikes the uncertainty he faces.

Consider a manager whose firm based wealth, noted by )(xf ,

consists of a call option on one share of the firm’s stock with a market price

noted as x that follows geometric Brownian motion:25

xdzxdtdx (A3)

where is the drift parameter, is the standard deviation of xdx / and dz

is a standard Wiener process with zero mean and variance of dt.

Let ))(( xfu represent the derived utility function, where

)()()( 2

)(

1 dXNedxNxf tTr , the Black/Scholes (BS) European option

pricer. )( 1dN is the normal cumulative evaluated at 1d and

1d is equal to

tT

tTrXxd

t

))(2/()/ln( 2

1 , )(1)( 12 dNdN ,r is the risk free

interest rate, X is the strike price, and (T-t) is the time to maturity.

The derived coefficient of absolute risk aversion is given as:

)(

)()(

fu

fufA

(A4)

Outright stock ownership has a delta of one. Thus, if total firm based

wealth is held as outright stock ownership, it is easy to verify that equation

(A4) reduces to equation (A2), the original Pratt measure.26

Ross (2004) has shown that the relationship between option based

incentives and corporate decision-making can be broken down into three

25 For expository convenience to identify the pertinent variables and their effect on risk

aversion, we consider only option based firm wealth. It is straightforward to incorporate

portfolios of individual options and outright shareholdings in the analysis because the pertinent

variables for the portfolios are linear combinations of the variables of the individual options

and shareholdings. 26The delta of total firm based wealth is the weighted average of the deltas of the individual

shareholdings and options: i

iiz , where i refers to the individual shareholdings and

options and iz is the proportion of asset i in total firm based wealth. For example, consider a

manager with 50% of his firm based wealth in shares and 50% in an option with a delta of 0.5.

The delta of his portfolio will be equal to 75.05.05.015.0 .

.




23

distinct effects: convexity, magnification and translation. The convexity effect

is defined as

Convexity effect = )(

)(

xf

xf

(A5)

In the Black/Scholes option pricer )(xf corresponds to delta and )(xf

corresponds to gamma where 1)(0 xf and )(0 xf . Since both are

positive, the convexity effect on managerial risk aversion is negative.27

To see how delta affects the convexity effect, let delta change while

holding gamma constant:28

)(

)(

)(

xfd

xf

xfd

0)(

)(2

xf

xf

(A6)

This derivative is positive because both )(xf and )(xf are positive. Thus,

an increase in delta will reduce the effects of convexity and increase

managerial risk aversion.

The magnification effect is defined as:

Magnification effect = 1)()( xffA (A7)

and since )(xf is positive and less than 1, the magnification effect on risk

aversion is also negative. The effect of a change in delta on magnification is

given by holding )(xf constant and letting delta change:29

)(

1)()(

xfd

xffAd0)( fA (A8)

This derivative is positive because )( fA is positive due to the risk aversion

postulated in equation (A1).

27 The Black-Scholes formula for the delta and gamma of a European call option are

respectively )()( 1dNxf and

21

2

)(

)(2

1)(

d

etTx

xf

.

28 For example, in the context of a simple call option this can be achieved by making

appropriate changes in the strike price and the option’s time to maturity. In the more general

context of a portfolio of options and shares the same effect can be achieved by adding an option

or portfolio of options or options and shares with a different delta but the same gamma. 29Again, this can be achieved by making appropriate changes in the strike price and the option’s

time to maturity.




24

The translation effect in Ross (2004) is given as:

Translation effect = )( fA - )(xA (A9)

This effect on risk aversion depends on whether the manager’s utility

function has decreasing, increasing or constant absolute risk aversion. Since

xf , the effect is positive with DARA.30

It is negative with IARA31

and it

is equal to zero with CARA.32

The translation effect depends on the level of

managerial firm based wealth and will change as the value of the firm’s shares

changes. Taking the derivative of (A9) with respect to x gives:

2

2

2

2

)(

)(

)(

)()(

)(

)(

)(

)( -

)()(

xu

xu

xu

xuxf

fu

fu

fu

fu

dx

xAfAd

(A10)

Both parentheses on the right hand side of equation (A9) have the same sign

with DARA and IARA. For DARA they are negative and for IARA they are

positive. Thus, since )(xf is positive, higher levels of delta increase the

absolute value of the first term on the right hand side of the equation and,

consequently, decrease the translation effect for utility functions with DARA

and IARA. It has no effect for utility functions with CARA.

Althoughvega does not appear directly as a determinant of managerial

risk aversion in the three effects outlined above, it could, however, affect risk

aversion indirectly through the effect of sigma on managerial wealth. It is well

known in the option pricing literature that 0 vega d

df.33

An increase in

30DARA implies 0)(

)( -

wu

wu

dw

dA

31IARA implies 0)(

)( -

wu

wu

dw

dA

32CARA implies 0)(

)( -

wu

wu

dw

dA

33 The Black-Scholes formula for the vega of a European call option is

2

)( 21

2

1d

etTxf

.




25

sigma increases the manager’s firm based wealth.34

The higher the vega, the

larger is the increase in wealth. An increase in wealth affects risk aversion

depending on whether the manager has decreasing, increasing or constant

absolute risk aversion. As with delta, the effect is negative with DARA,

positive with IARA and equal to zero with CARA.

References

Aboody, D., Hughes, J., Liu, J., 2005. Earnings quality, insider trading, and

cost ofcapital. Journal of Accounting Research 43, 651–673.

Abdel-Khalik, A. R. (2007), An empirical analysis of CEO risk aversion and

the propensity to smooth earnings volatility,Journal of Accounting,

Auditing and Finance, 22 (2), 201-235.

Armstrong, C., A. Jagolinzer, and D. Larcker, 2010, Chief executive officer

equity incentivesand accounting irregularities, Journal of Accounting

Research 48 (2), 225-271.

Armstrong, C.S., Larcker, G.F., Ormazabal, G., and Taylor, D.J. (2013), The

relation between equity incentives and misreporting: The role of risk-

taking incentives, Journal of Financial Economics, 109, 327-350.

Becker, B. (2006), Wealth and executive compensation. The Journal of

Finance 61, 379–397.

Belghitar Y, Clark E, Kassimatis K (2011), The prudential effect of strategic

institutional ownership on stock performance.International Review of

Financial Analysis, 20 (4) 191-199.

Belghitar, Y. and Clark, E (2014) Theeffect of option based managerial

compensation on corporate cash holdings. Journal of Financial

Research,37 (2) 191-210.

Bell, L. and T. Jenkinson 2002. New evidence of the impact of dividend

taxation and on the identity of the marginal investor, Journal of

Finance 57,1321-1346.

Bergstresser D., Philippon T., 2006. CEO incentives and earnings

management. Journal of Financial Economics 80, 511-529.

34Other things being equal, the vega of a share is equal to zero. Like delta, the vega of

managerial total wealth is a weighted average of the vegas of the individual shareholdings and

options: i

ii vegazvega .

http://dx.doi.org/10.1016/j.irfa.2011.04.002

http://dx.doi.org/10.1016/j.irfa.2011.04.002




26

Black F., Scholes M., 1973.The pricing of options and corporate liabilities.

Journal of Political Economy (May-June), 637-654.

Brockman P., Martin X., Unlu E., 2010. Executive Compensation and the

Maturity Structure of Corporate Debt. Journal of Finance 65, 1123-

1161.

Brown, S., Lo, K., Lys, T., 1999. Use of R2 in accounting research: Measuring

changes in value relevance over the last four decades. Journal of

Accounting & Economics 28, 83–115.

Burgstahler, D., Dichev I., 1997.Earnings management to avoid earnings

decreases and losses. Journal of Accounting and Economics 24, 99-

126.

Burns, N.,Kedia, S., 2006. The impact of performance based compensation on

mis-reporting. Journal of Financial Economics 79, 35-67.

Caramanis C., Lennox C., 2008.Audit effort and earnings management.

Journal of Accounting and Economics 45, 116-138.

Chava, S., Purnanandam, A., 2010. CEOs versus CFOs: incentives and

corporate policies. Journal of Financial Economics, 97, 263–278.

Childs, P. D., Mauer, D. C., Ott, S. H., 2005. Interactions of corporate

financing and investment decisions:The effects of agency conflicts.

Journal of Financial Economics 76, 667-690.

Cohen, D. A., Dey, A. and Lys, T. Z.,2008.Real and accrual‐based earnings

management in the pre‐ and post‐Sarbanes‐Oxley periods. The

Accounting Review 83, 757-787.

Coles J.L., Daniel N.D., Naveen L., 2006. Managerial incentives and risk-

taking. Journal of Financial Economics 79, 431-468.

Cornett, M., Marcus, A., Tehranian, H., 2008. Corporate governance and pay-

for-performance: the impact of earnings management. Journal of

Financial Economics 87, 357–373.

Daniel, N. D., J. Martin J.S.., and Naveen, L., 2004, The hidden cost of

managerial incentives: evidence from the bond and stock markets,

Working Paper, Arizona State University.

Dechow P.M., Sloan R.G., 1995. Detecting Earnings

Management.Accounting Review 70, 193-225.

DeFond, M.L., Park, C.W., 1997. Smoothing income in anticipation of future

earnings.Journal of Accounting and Economics 23, 115–139.

Erickson, M. Hanlon, M., and Maydew E., 2006, Is there a link between

executive equity

incentives and accounting fraud? Journal of Accounting Research 44, 113-143.




27

Fama E.F. , 1980. Agency Problems and the Theory of the Firm. Journal of

Political Economy 88, 288-307.

Feng, M., Ge, W., Luo, S., Shevlin, T., 2011. Why do CFOs become involved

in material accounting manipulations? Journal of Accounting and

Economics 51, 21–36.

Francis, J., LaFond, R., Olsson, P., Schipper, K., 2005. The market pricing of

accrualsquality. Journal of Accounting and Economics 39, 295–327.

Gaver J.J., Gaver K.M., Austin J.R., 1995. Additional evidence on bonus

plans and income management. Journal of Accounting and Economics

19, 3-28.

Grant J., Markarian G., Parbonetti A., 2009. CEO Risk-Related Incentives and

Income Smoothing. Contemporary Accounting Research 26, 1029-

1065.

Guay W.R. , 1999. The sensitivity of CEO wealth to equity risk: an analysis

of the magnitude and determinants. Journal of Financial Economics

53, 43-71.

Harris M., Raviv A., 1990. Capital Structure and the Informational Role of

Debt. Journal of Finance 45, 321-349.

Healy, Paul M. and J.M. Wahlen, 1999, A review of the earnings management

literature and its implications for standard setting, Accounting

Horizons13, 365-383.

Huang, P., Zhang, Y.,Deis, D.S., and Moffitt,J.S. 2009. Do artificial income

smoothing and real income smoothing contribute to firm value

equivalently?, Journal of Banking & Finance 33, 378-4266.

Hull, J.C., 2003. Options, futures and other derivatives.Prentice Hall, USA.

Hutchinson M., Gul F.A., 2004. Investment opportunity set, corporate

governance practices and firm performance. Journal of Corporate

Finance 10, 595-614.

Jensen M.C., Meckling W.H., 1976. Theory of the firm: Managerial behavior,

agency costs and ownership structure. Journal of Financial Economics

3, 305-360.

Jiang, J., Petroni, K., Wang, I., 2010. CFOs and CEOs: who have the most

influence on earnings management? Journal of Financial Economics

96, 513–526.

Klein A. , 2002. Audit committee, board of director characteristics, and

earnings management. Journal of Accounting and Economics 33, 375-

400.




28

Knopf J.D., Nam J., Thornton J.,John H., 2002. The Volatility and Price

Sensitivities of Managerial Stock Option Portfolios and Corporate

Hedging. Journal of Finance 57, 801-813.

Kothari, S., Loutskina, E., Nikolaev, V., 2011, Agency theory of

overvaluedequity as an explanationfor the accrualanomaly, Working

paper, MIT Sloan School of Management.

Kothari S.P, Leone A.J., Wasley C.E., 2005. Performance matched

discretionaryaccrual measures. Journal of Accounting and Economics

39, 163-197.

Lambert. R. A. 1986. Executive effort and the selection of risky projects.Rand

Journal of

Economics 17, 77-88.

Liu, Y., and Mauer, D., 2010. Corporate cash holdings and CEO

compensation incentives. Journal of Financial Economics, Article in

Press.

Low,A., 2009.Managerialrisk-taking behaviour and equity-based

compensation.JournalofFinancialEconomics92,470–490.

Murphy, K., 1999. Executive compensation. In: Ashenfelter, O., Card, D.

(Eds.), Handbook of LaborEconomics, vol. 3b (Chapter 38). Elsevier

Science, North Holland, 2485–2563.

Ozkan A., Ozkan N., 2004. Corporate cash holdings: An empirical

investigation of UK companies. Journal of Banking & Finance 28,

2103-2134.

Peasnell K.V., Pope P.F., Young S., 2005. Board Monitoring and Earnings

Management: Do Outside Directors Influence Abnormal Accruals?

Journal of Business Finance & Accounting 32, 1311-1346.

Perry, T., Zenner, M., 2000. CEO compensation in the 1990s: shareholder

alignment or shareholderexpropriation? Wake Forest Law Review 35,

123–152.

Pratt J.W. , 1964. Risk Aversion in the Small and in the Large. Econometrica

32, pp. 122-136.

Rees, L., Gill, S., Gore, R., 1996. An investigation of asset write-downs and

concurrent abnormal accruals. Journal of Accounting Research 34,

157–169.

Ross, S. A., 2004.Compensation, incentives, and the duality of risk aversion

and riskiness. Journal of Finance 59, 207-225.




29

Shin H., Stulz R.M., 2000. Firm Value, Risk, and Growth

Opportunities.NBER Working Paper 7808, National Bureau of

Economic Research, Inc.

Shleifer A., Vishny R.W., 1992. Liquidation Values and Debt Capacity: A

Market Equilibrium Approach. Journal of Finance 47, 1343-1366.

Sloan, R., 1996, “Do Stock Prices Fully Reflect Information in Accruals and

Cash Flows about Future Earnings?” Accounting Review, 71, 289

Smith C.W., Watts R.L., 1992. The investment opportunity set and corporate

financing, dividend, and compensation policies. Journal of Financial

Economics 32, 263-292.

Stulz R.M. , 1984. Optimal Hedging Policies. Journal of Financial &

Quantitative Analysis 19, 127-140.

Stulz, R.M. and Smith, C.W. 1985. The determinants of firms’ hedging

policies. Journal of Financial and Quantitative Analysis 28, 397-405.

Teoh, S., Welch, I., Wong, T., 1998a.Earnings management and the long-run

underperformance of seasoned equityofferings. Journal of Financial

Economics 50, 63–100.

Teoh, S., Welch, I., Wong, T., 1998b.Earnings management and the long-run

underperformance of initial public offerings. Journal of Finance 53,

1935–1974.

Titman S., Wessels R., 1988.The Determinants of Capital Structure Choice.

Journal of Finance 43, 1-19.

Tong Z. , 2010. CEO Risk Incentives and Corporate Cash Holdings.Journal of

Business Finance & Accounting 37, 1248-1280.

Weir C., Laing D., McKnight P.J., 2002. Internal and External Governance

Mechanisms: Their Impact on the Performance of Large UK Public

Companies. Journal of Business Finance & Accounting 29, 579.

Date post:	11-Mar-2018
Category:	Documents
Upload:	hanhan
View:	220 times
Download:	4 times

Firm Organizational Environment and The Effect of Options ... · PDF...

Documents